Biologic Specialization in the Genus Septoria

BIOLOGIC SPECIALIZATION IN THE GENUS SEPTORIA

WALTER SPURGEON BEACH

B. S. University of Minnesota, 1914.

M. S. Michigan Agricultural College, 1915.

THESIS

Submitted in Partial Fulfillment of the Requirements for the

Degree of

DOCTOR OF PHILOSOPHY

IN BOTANY

THE GRADUATE SCHOOL

OF THE

UNIVERSITY OF ILLINOIS

1918 Digitized by the Internet Archive

in 2013

http://archive.org/details/biologicspecialiOObeac UNIVERSITY OF ILLINOIS

THE GRADUATE SCHOOL

DmT X 19lX

I HEREBY RECOMMEND THAT THE THESIS PREPARED UNDER MY

SUPERVISION BY.

ENTITLED.

BE ACCEPTED AS FULFILLING THIS PART OF THE REQUIREMENTS FOR

In Charge of Thesis

Head of Department

Recommendation concurred in*

Committee

Final Examination*

*Requireci for doctor's degree but not for master's CWJC Table of Contents

Page I. Introduction ------l

II. Historical ------4

III. Experimental Methods and Material ------9

IV. Septoria polygonorum Desra. ------12

V. Septoria lactucicola E.& M. ------16

VI. Septoria lactucae Pass. ------19

VII. Septoria tritici Desra. ------23

VIII. Septoria malvicola E.& M. ------30

IX. Septoria scrophulariae Peck ------32

X. Septoria convolvuli Desra., Septoria septulata sp.nov. 33

XI. Septoria verbascicola B.& C. ------37

XII. Septoria cirsii Niessl. ------40

XIII. Septoria brunellae E.& H. ------42

XIV. Septoria lycopersici Speg. ------43

XV. Septoria lepidiicola E.& M. ------45

XVI. Septoria helianthi E11.& Kell. -- -47

XVII. Septoria rubi West. ------51

XVIII. Septoria atro-purpurea Peck ------52

XIX. General discussion

Age incidence ------53

Susceptibility of different leaf surfaces ----- 53

Effect of the mass of inoculum ------54

Effect of wounding ------54

Variations in the morphology of the fungus - - - - 55

Host limitations ------56

The value of disease characters ------57

Biologic specialization ------7

Page XX. Conclusions 5^

Literature Cited 60

Graph I with Explanation ...... --63

Plate I with explanation ...... 55 Biographical Sketch ------6$

PIOLOCT" 9P7riAT.IZATTO\T T v TH^ GENUS SFPTOPIA

T Introduction

The genus ?eptoria presents an excellent field for the

studv of biologic specialization on account of the very large number ar.d the uncertain status of the described species, cf which there are more than 1200. The morphological characters for the

limitation of these are few, consisting chiefly of spore shape,

length, thickness, and septation, and in some instances a slight

tinting of the spore; pycnidium sh»r*e, size, and color; and os-

tiole size and character. The ciclc^ical characters for the lim-

itation of species are the size, shape, margin, conation, and color cf the disease spots, and their location and distribution upon the

host. The specific descriptions in many cases are meager and real-

ly useless. The presence of a Peptoria upon an unrecorded host is

often made the basis of a new species.

By tabulating (10), according to spore length, all the "spe-

cies" of Septoria given in SaccardO 1 a "Sylloge Fungorum", it has been found that nearly 700 fall within the limits of 20 to 50 mi-

crons. Twenty-six species on grasses are between 20 to 40 microns.

Yet in one esse a Saccardian description gives a spore length rant?-

insr from 19 to 62 microns. There have been sreat inaccuracies in measuring spores and pyenidia., with tendencies tc report the measurements in round numbers. Elliott (7) has shc*.vn the degree of

personal error with our modern technique. Among eleven observers

measuring many spores of Alternaria, he found that there was a

variation of over 41$ from the highest measurement returned for

maximum length, and that such a variation would make it impossible

to distinguish the species of Alternaria on the basis of spore

measurement. Septation is largely dependent upon the stage of ma-

turity of the spores, is often difficult to see, and was ignored

by early mycologists. Enviromental conditions cause fluctuations

in the size of spore9 and pycnidia. The value of disease charac-

ters is doubtful for they commonly change with the age, the part,

and the species of the host, as well as with weather conditions.

The confusion that Cavara (4) met regarding these points in his at-

tempt to distinguish Septcria t rit ici and S. graminum on wheat, led

him to state that both forms probably belonged to the same morpho-

logical species.

Little is known respecting the existence of biologic special-

ization in the genus Septcria. Except for a few instances (15,

22, 45) no cross-inoculations have been made to ascertain whether a

Septoria, as found upon a particular host, is restricted to that host, or is more or less cosmopolitan and adaptive in its parasitism.

In the literature at present one is confronted by two extremes of

treatment: (a) closely similar species are described on the same

host, e.g. , Sept or ia lactucae and S. consimili s on Lact uca sat iva;

(b) more widely variant forms on distantly related hosts are de-

scribed as representing one species, e.g., S. graminum on twelve

or more genera of grasses, and a species of sedfre. It is obvious

that such pra.ctice is not consistent, and cannot adequately reflect

the truth.

The facts as presented above are illustrative of the present

unsatisfactory status of the species not only of Septcria but also

3 of many other genera of the Fungi Imperfecti. More accurate knowledge regarding host characters, morphological variation, and biological specialization is greatly needed from the standpoint of classification. The genus is the more worthy of studv on account of its high economic importance. To study such a large genus as

Septoria in culture, either upon the living hosts or upon artificial media, would require the time of many investigators. The present paper is intended merely as a contribution to the general problem as suggested above, and the investigations were conducted with the following leading objects:

1. to determine the host range of as many species of Septoria as possible in order to ascertain; (a) whether morphologically similar forms from different hosts vary in host range, (b) whether morphologically unlike forms ever have the same host range; or in other words to ascertain whether anv of the species available for study consist of a number of biological forms, and whether any now listed as distinct species are identical.

2. to compare disease characters, i.e., the host response to infection, when produced (a) by a single species of Septoria upon dissimilar hosts, or (b) by different species of Septoria on the same host.

3. to note any morphological changes in the size and shape of spores and pycnidia as a result of change of host or of the condition of the host, or of other environmental factors.

4. to determine the susceptibility of different parts of the

same host and of these at different stages of maturity, and to note whether any distinctions in disease characters are correlated with host structure and condition.

II Historical

The literature dealing with biologic specialization is extensive. A comprehensive review cannot be undertaken here, yet a brief summary of the essential points, especially as brought out bv the investigations of rusts and powdery mildews, will be useful in the interpretation of results to be presented. It may be well to mention thRt biologic specialization is also spoken of as "adaptive parasitism", and biologic forms as "biologic species", "biologic races", "physiological races", and "adapted races. "

While Friksson was first to conduct extensive experimental work in biologic specialization, the idea was in the mind3 of earli er investigators for Schro'ter (8, footnote) in 1379 spoke of split- ting up the old species Puccinia caricis into a number of forms differing slightly morphologically, but widely in biologic charac- t eri sties.

Eriksson (8) called attention more definitely to the nature of biologic specialization through the results of his cross- inoculation experiments with Puccinia gramlnis in Sweden. He found that, although this rust upon the cereals and grasses represented the same morphological species, the form upon one host-species was not always identical with the form upon another; since, for example oat rust would transfer to oats, but not as a rule to other species of Gramineae. He showed that this fungus embraced at least six distinct forms distinguished by their dissimilar powers of infection with respect to the species of the grass family. In a subsequent paper (9) Friksson showed that the trend of specialization may be different in isolated localities. This wa3 illustrated by

the fact that the form on rye, Puccinia graminis secalis . has a

5 relatively vigorous development in Sweden, but a relatively weak one in North America. In the case of Puocinia graminis trit ici on wheat the more vigorous development is in the latter country.

Hitchcock and Carleton (13) performed early infection experiments with Puccinia gramini3 in the United States, and obtained results in agreement with those of Eriksson. Carleton (2,3) in working with biologic forms of rusts has secured valuable data upon varietal resistance in cereals.

Ward (46), experimenting with Puccinia dispersa upon Promus, showed the presence of several biologic forms of this fungus on this single host-genus, and discovered "bridging species". The nature of a "bridging species" is described by this statement of the author: "although it is generally true that the adapted races of

Puccinia dispersa are restricted to groups of closely allied species, there do occur species which serve as intermediaries in the passage of the fungus from one section of the genus to the other. "

Arthur (l) reported that Puccinia helianthi is divided into a number of biologic races, and that Helianthus annuus acts as a bridging host,

Stakman (42) in a study of the physiological races of rusts showed that the power of wheat rust to infect rye was somewhat in- creased when the fungus lived a number of generations on barley.

Recently Stakman and Piemeisel (43) have published the results of extensive infection experiments with Puccinia graminis from 35 species of native grasses collected in widely separated regions of the Northwestern States. More than one biologic form was found to occur on the same host in nature. It was necessary

6 to employ differential hosts to determine the identity of the for^s. These authors claim, moreover, that the biologic forms of

Puccini a grami nis can be distinguished from one another morphologically as well as biologically, by the size, shape, and color of the urediniospores. Regarding susceptibility they state, "all gradations in susceptibility occur from complete immunity to complete susceptibility to the various biologic forms. The following reactions mav be made to inoculation: no visible effect, appearance of small flecks, production of very small uredinia without flecks, production of very small uredinia in small or large flecks, production of large uredinia surrounded by small dead areas or by apparently healthy tissue.

With the ^rysiphaceae the extension of our knowledge of biologic specialization has had a similar course, Although Salmon had discussed the probable existence of biologic specialization in this group of fungi, Neger (21) was first to report definite data. Neger said that Frysiohe cicho rac earum would not cross- infect among certain of the numerous hosts of this mildew.

Marchal (l?) soon followed Neger with a paper describing

seven "formes speciali sees" in the Oidium of Erysiphe graminis . and said each ms confined to a single genus of grasses or cereals.

Salmon t v en attacked the problem of adaptive parasitism in the powdery mildews, and published a. long series of papers (1903-

1906). His important work concerns the Oidia on Promus and

Hordeum. He proved that several biologic forms of the Oidium. parasitized the species of Bromus, that an individual species may be the meeting place of several biologic for^s, and that "bridging species" existed. An additiona.l discovers is best stated in his own words, (35, p 57) "the inter-relations of the biologic forms with certain of their host-plants become complicated by the existence of

'biologic forms' of the host plants." Salmon found that in Promus mollis there was a susceptible and an immune race with respect to four separate biologic forms of the mildew. In hi3 study of the various kinds of cultivated barleys (32), he was able to arrange the varieties in a series according to their susceptibility. In experiments with ascospores (31,37) he found that adaptation in the perfect 3tages of Erysiphaceae is parallel to that in the conidial.

Salmon made histological studies (38) to see what development the mildew reached when sown upon non-susceptible leaves. He discovered that the germ tube of the conidium penetrated the epidermal cell and sometimes formed an incipient haustorium within, but that further development was usually arrested, indicating that resistance was a physiological factor of the cell contents, and not morphological. He shewed (39) that resistance to a biologic form can be overcome by injuring the host, or by treating it with certain nar- cotic vapors.

G. M. Peed has been the chief worker in the United States upon adaptive parasitism in the powdery mildews, and has published numerous papers (1905-1916). In addition to corroborating- the results of

Salmon and others, he has carried out extensive croes-inoeulat ion experiments with Frysiphe g rami, pis (27,28,29) in which he has shown considerable variation in susceptibility among the species and varieties of Triticum, Hordeum, Avena, and Secale, a few of which are immune. In Frys iphe cjiicpracearum (25) on Cucurbitaceae he found but one biologic form, capable of growing upon various members of this host-family.

The more important instances of biologic specialization reported in other groups of fungi are as fellows: by Stager in

Claviceps (41), by Diedicke in Pleospora and its conidial stage

Helminthosporium (5), by Gilbert in Plowrightia morbosa (11), by

Miiller in Rhy t i srr.a acerinum (SO), and by Hesler in Sphaercpsis malorum (12). Shear and Wood (40) found that the races of

Olomerella cingulata, from different hosts varied sorrewhat in the vigor of their attack upon other hosts. Tests by Westerdijk (47) indicate the absence of biologic specialization among similar races of Scl erotinia libert iana. Pands (23) has proven bv cross- inoculation that the Alternarias of Datura and potato are not identical as some have supposed.

With species of Septoria only a few investigators have worked. Levin (15) inoculated several plants akin to the tomato with Sep toria lycopersici. Definite small black spots without pyenidia appeared on potato, but no effect was seen in other plants. Norton (22) performed similar experiments in humid inclosures, and obtained infection upon several species of Solarium.

A further discussion of this author's results will be given in a subsequent section. Stone (45) proved by infection that Septoria ribes and likewise its perfect staere Myc03phaerella groasulariae taken from Ribes nigrum will infect R. gros sularia, P. rubrum, and

P. Qj^yacanthpide8^

Montemartini (18) has concluded that parasitic fungi are ex- tremely sensitive to the chemical composition of the nutritive medium on which they live, and under its influence acquire characters of adaptation which attain to a certain fixity. In conse- quence fungi may become unable to flourish on species different

9 from those to which they have accustomed themselves, or even on other portions of the same plant which they inhabit, or in different developmental stages of such plant or organ. This problem is further complicated by atmospheric conditions, and the influence there- of upon sensitivity to attack as well as upon the virulence of the infecting bodies.

The review as here given serves to illustrate the intricate nature of biologic specialization, and to remind one of the various factors to be considered in the interpretation of the results of inoculation experiments.

Ill Experimental Methods and Material

The species of Septoria for investigation were with few ex- ceptions collected in the vicinity of Urbana, Illinois. During the growing season spore-bearing material was gathered in the field from naturally infected plants. Leaves with disease spots were usually placed in a moist chamber for a day to cause heavy sporu- lation. In fall and winter spores were obtained in most cases from pure cultures isolated before the first frost. None of the cultures appeared to lose virulence, and it was found that this was the most certain way to keep infection material in readiness. In preparation for inoculation, spore suspensions were made with dis- tilled water and were concentrated enough to contain fifty to one hundred spores per loop.

The experiments were begun in June and were continued to the following April. In summer the work was conducted in the field upon plants growing in their natural state. Freedom from disturbance and as far as possible from natural infection governed the selection

10 of suitable individuals. After the first frost the experiments were continued in the greenhouse upon potted plants transplanted from nature or grown from seed.

The method of inoculation was somewhat varied. In the field an atomizer was most convenient for the spore suspension could be carried safely in the glass container. In the greenhouse a wire loop was mere commonly used to spread drops of the suspension over the leaf surface. Where a bloom or other epidermal structures made it difficult to secure good contact, especially with leaves of grasses, the suspension was rubbed on with the clean finger tips with proper care. Consistency of method was followed in a single series to secure comparable results. The entire surface of certain leaves was inoculated, while adjacent lea.ves of the same plant or near plants were used as checks to detect foreign infection where there was danger of this. Inoculations were made upon individuals of the original host to afford a check upon the viability of spores, or upon unfavorable environmental conditions.

Some difficulty was experienced in providing conditions favorable for infection. At first bell jars were placed over the plants for the customary period of two days, but later it was found better to vary this from three to five days. Paraffined paper bags were subsequently substituted for bell jars as they were more convenient to handle in the field, and gave more uniform results, po33ibly because they provided partial shade. The use of these bags was continued in the greenhouse. In cases in which disease spots without pyenidia resulted, the bags were replaced over the plants for a second period to promote the growth of the fungus, and to induce if possible the development of spore-bearing bodies. If this failed

11 the spotted leaves were detached and put into a sterile moist chamber. In this manner pycnidia were obtained and infection proven in

some cases in which results would otherwise have been in doubt.

Thi3 was especially necessary in instances in which spore suspen-

sions were made from natural material, and the spores of other fungi might have been present to cause the spots.

In the following accounts the results of the inter-inoculations with each species of Septoria are presented graphically by means of diagrams, and in more detail by means of tables. The ar-

rows in the diagrams indicate infection, and the lines ending in bars non-infection. The denominators of the adjacent fractions

state the total number of leaves inoculated, and the numerators the number of leaves having one or more infected spots. The total number of spots formed is given in the tables where this is required for a closer comparison of susceptibility. In column two of the tables the letter f indicates that the experiment was in the field or out of doors, and £ in the greenhouse. The letter b with the accompanying figure records the number of days the plant was bagged

immediately following inoculation.

The lists of hosts given were compiled partly from Saccardo's

"Sylloge Fungorum", and partly from data furnished by the herbariums

at the New York Botanical Garden , and in the United States Depart- ment of Agriculture.

IV Septorla polygonorum Desm .

All the collections of Septoria made from species of Polygonum

at Urbana were determined as Septo ria polygonorum Desm . This fungus has been reported upon the following species of Polygonum, which are arranged in groups:

A. Infected by author B. Inoculated but not infected

P. persicaria P. amphibium H pennsylvanicum " hydropiper n lapathifolium " convolvulus w orientale

C. Unused in experiments D. Synonymous

P. bi3torta P. incarnatum = P. lapathifolium sieboldii " nodosum = w n " mitis hydropiperoides " muhlenbergii M dumetorum

The members of group C are foreign, or were not found growing in thi3 vicinity. The species of group B, as well as others that gave negative results when inoculated, apparently had sufficient opportunity to become infected in nature; for infected plants of smartweed were usually growing in close proximity to the uninfected kinds. Yet repeated search for material resulted in finding Sep -

toria polygonorum upon only Polygonum persicaria , P. pennsylvanicum ,

and P. lapathifolium , upon which the form in this locality appears to be specialized. The plants of P. orientale were few and apparently escaped infection by being isolated. It remains an open question

the forms of S_. olygonorum P. P. whether p reported upon amphibium , hydrooiper , and P. convolvulus of group B are differently specialized or whether the hosts are resistant in this region. The former case is likely if the fungi and their hosts were correctly determined.

With respect to the power of the fungus on one species of group A to infect the other susceptible species, there appears no certain distinction air o Tig the Sept aria forms from these hosts, at least in the light of the limited number of experiments. On July

10 (table I) a plant each of P. pennsylvanicum and P. persicaria growing together in a shaded place was inoculated with spores from

P. persicaria . Heavy infection resulted in the latter, while P. pennsylvanicum remained uninfected. Had there been previous natural infection, it is very improbable that the plant of P. pennsylvanicum should have escaped it, for the branches of the two specimens almost interlocked. Thus it seems that this individual of

P. pennsylvanicum was resistant. On Oct. 3 this experiment was repeated upon a plant of P. pennsylvanicum that had been watched under a bell jar for previous infection. This specimen pave full

infection with spores from P. persicaria .

Septoria polygonorum was isolated, and the growth characters were alike in the fungus a3 isolated from each of the four hosts in group A.

Although a vigorous parasite upon the few species of Poly- gonum, the form of Septoria polygonorum dealt with here is not at all cosmopolitan in its parasitism. A special attempt was made to infect a plant of P. convolvulus kept constantly covered with a paraffined paper bag to insure a humid atmosphere and provide partial shading, but with negative results. One hundred leaves of P. hydropiper also save negative results. The fungus was induced to attack detached leaves of P. convolvulus kept in a moist chamber in the warm laboratory, spots and pycnidia developing in seven days; but it would not attack Fumex crispus or Fagopyrum esculent urn in a

14 Table I

Infections with Septoria polygone rum Desrr.

Condi- No. of Check in- Date tions Plants inoculated leaves fections on infected & No. of original Spores from Polygonum pennsylvanicum host

June 36 f-b 2 Polygonum persicaria 12/16 many 8/10 Aug. 30 f-b 2 N 11 11/20 many n n Sept. 6 f-b 3 Offt/ Xo1 p Sept. 14 f-b 3 it it 10/15 26 *rX/ CO

Aug. 30 f-b 4 Polygonum lapathif olium 3/10 34 Sept. 14 f-b 4 it n 5/5 many 6/6 ft/1 K

O a. vs -A- T A oept. 14 f-b 4 Polygonum orientale fi/&D/ O pft 6/6

June 9 shaded Polygonum erectum o/io 4/5 June 28 f-b 3 0A1 8/10 0/21 12/15

June <5o f-b Polygonum convolvulus 0/8 8/10

June fib shaded Polygonum hydropiper 0/4^ 8/10

Sept. 7 shaded Polygonum acre U/ la Sept. 21 f-b 4 ft n 0/20 p}us 0/32 bept. cl shaded Polygonum scandens 0/25 plus

June 28 shaded Polygonum amphibium 0/7 8/10

Oopuy*>. r\ r~v* 63a ^ from Polygonum persicaria

July 10 shaded Polygonum pennsylvanicum w/ J. V./ 10/15 Oct. 3 n it 5/5 many 5/15 oept. o f-b 4 Polygonum lapathifolium 7/") O mciiiy 10/10 oept.

July 10 shaded Polygonum convolvulus 0/20 10/15 Mar. 3 g-b 3 n n 0/5 0/25

July 10 shaded Polygonum hydropiper 0/30 10/15 Sept. 28 f-b 4 it it 0/25 " plus 0/55

July 10 f-b 3 Polygonum erectum 0/25 10/15

.. .

Table 1 continued Infections with Septori a polygonarum Desm,

0/10 Sept . 7 shaded Polygonum acre Sept .22 f-b 4 W N 0/30

S pores from Polygonum lapathif olium

"U 1 — Sept .11 I - D O jrOJ.y gonum pt; ilxl ay X V etii 2/5 15 icura

Sept .21 f-b 3 Polygonum persicaria 6/12 15 8/8 Mar 10 g-b 3 M N 3/4 13 9/16 28 many 8/8 Sept .21 I — E? O Polygonum orientale 7/7

Sept .21 f-b 4 Polygonum hydropiper 0/25 8/8 8/8 Sept .21 f-b 4 1 >ly gonum acre 0/20

r\ /on Sept .21 shaded Polygonum scandens O/ o

Oct. 7 g-b 4 Polygonum convolvulus 0/12

Dec 3 moist Polygonum convolvulus 3/7 chamber (leaves detached) Oct. 7 g-b 4 Fagopyrum esculentura 0/10

Dec wT moist Fagopyrum esculent urn 0/7 chamber (leaves detached) Dec. O1 moist Rumex crispus 0/7 chamber (leaves detached)

Diagram I Infection with Septoria polygonorum from Poly- gum pennsylvaiiicum, P. persicaria, and P. lapathif oliura

16 similar way. Thus Septoria pelvgonorum appears ag a species with fairly fixed infection-powers, possibly embracing forms differently specialized in addition to the one experimented with here.

The disease spots are not the same on the various species of

Polvgonum. The distinctions are illustrated in Plate I, Figs. 1-4.

The specimens were selected as representing the usual character of the spct3. Specimens of the same species might have been selected shewing scarcely distinguishable spots. Upon P. p enn s yl van i cum the spots attain the greatest size, often become irregular in outline, have rut a very narrow, dark-brown border, and have pyenidia uni- fcrmlv distributed throughout their area (Fig. 1). The 3pcts upon P. psrsicaria are intermediate in 3ize, and are chiefly characterized by a rather wide, dark reddish-brown border. The spots seldom lose their orbicular shape, and the pyenidia are grouped nearer the center (Fig. 2). Upon P. lapath i_ folium the spots range smallest in size, and are yellowish-brown in the fresh leaves, with a border of the same color, but in the dry specimens the spots become reddi sh-brewn

(Fig. 3). In 3ize and shape the spots on P. orientals resemble those on P. pennsyl vani cum, but are yellowish-brown in the fresh leaf. The reddish-brown border appeared when the leaf was put into a moist chamber for a day. The variability in the biological or host characters shown here illustrates the untrustworthy nature of these characters when used to distinguish species.

v Septoria lactuciccla F. & M.

The plants inoculated with fjptcria lactuciccla were grown in pots in the greenhouse. Some ha.d been ?rown from seed, others had

17 be^n transplanted from the field. A3 a rule only young or recently mature leaves were sown with 3pores. The spore material was ob-

tained from the field.

Sep to ria lactucicola ha3 been reported upon Lactuo a f 1c ri-

dana, L. acaricla, and L. canadensis, but in most instances on the

last plant. The results in Table II 3how that this ? epic ria is somewhat adaptive in its parasitism, being- able to attack three genera

and five species. Were the experiments conducted on a broa.der scale,

the best range would probably be enlarged. There is a gradation in

susceptibility passing from L. c anad ensis through to Sorchus olera.-

ceus. Although the fungus atta.cked L. sat iva and L. scariola rather

readily, with lc spots on 20 leaves and 10 spots on 35 leaves re-

spectively, the infection was slisrht compared with 23 spots on 7

leaves cf the original host. Moreover, there were but few fruiting

bodies on these new hosts. The most of the 17 spots on Prena.nthes

sp_. were very small, and Son onus oleraceus was a very uncongenial

plant for the fungus. The comparison will be more clear with a de-

scription of the spots.

Sep toria lactucicola produces upon L. canadensi s round, red-

dish-brown to black spots, usually from 3 to 15 mm. in diameter, and

frequently with concentric zones of lighter and da.rker color. L.

ac f ricla was the only new host in which these characters were pre-

served, and here the spots were lighter colored, probably on ac-

count of the thinner leaves. Upon L. sa,t iva the spots were spread-

ing, with an indefinite border, and several time3 larger than those

on L. canadensi 3 wher = the border is very definite. They were col-

ored light brown in the center, gradually changing to yellow sni

then to the green cf the normal tissue. No pyenidia were observed

18 Table 1

Infections with Septoria lactucicola E. & If.

from Lactuca canadensis

No. of leaves Condi- iniected & NO. Of Date tions Plants inoculated inoculated Spot 8 Remarks

Oct. 2 g-b 4 Lactuca canadensis 4/6 many Check Oct. 26 g-b 4 n it 5/ 7 23 n 9/13

Oct. 26 g-b 4 Lactuca sativa 10/30 16 Pycnidia few

Oct. 26 g-b 4 Lactuca scariola b/ oo 1U Pycnidia few

Oct. 2 g-b 3 Prenanthes sp. 2/8 3 Spots small Oct. 26 g-b 4 n n 1/10 4 it n Nov. 15 g-b 7 it it 4/8 n it 7/26 17

Julyll g-b 4 Sonchus oleraceus 2/6 3 Pycnidia in Aug. 30 g-b 3 n n 1/4 2 moist cham- Oct. 2 g-b 4 it tt 2/5 2 ber 5/15 7

Oct. 2 g-b 4 Sonchus asper 0/25 Oct. 26 g-b 3 n n U/ 4U 0/65

Oct. 2 4 Sonchus uliginosus 0/3 Oct. 26 g-b 4 n it 0/8

Oct. 2 g-b 5 Taraxacum officinale 0/8 Oct. 26 g-b 5 it n 0/14 0/22

s s

19 in the greenhouse, but they developed in two davs upon leaves detached and put into a moist chamber. For a further comparison of the spots on these first three hosts compare Plate I, ^iTs. 7-9. The spots upon Prenanthes Bp. were 1 mm. to 3 mm. in diameter, dark brown, more or less angular, and usually surrounded by a greenish-vellow zone. Pycnidia were 3een upon the plant, but spores developed onlv in the moist chamber. The spots upon Sonchus ole ra.ceu s were 1 mm. to

5 mm. in dianeter, grayish brown, and more or less limited bv the leaf veins. A small number of pycnidia with spores characteristic

°f Septoria lactucicola were obtained upon two of several infected leaves placed in a moist chamber.

That the biologic characters change with the host is clearlv shown. Such variation has doubtless led to erroneous det erminat ior or to the useless multiplication of species in rranv instances. En- vironment mav likewise cause alterations of these characters for the spots on Ij. sana densi tenaed to lose their concentric zones and become spreading; if kept long in a moist atmosphere.

VI Septoria lact uca e Pass.

The following- hosts are reported for Sept oria lactucae :

Lact uca sat iva , L. scariola, L. virosa , L. canadensis , and Chon- drilla ^ura.1^3. The fungus appears to cross readily between L.

sat iva and L. scariola , and causes upon both of these hosts marked disease. The passage of the fungus from either of these host3 to

L. canadensis apparently takes place onlv under very favorable conditions of heat and moisture, for it wa.s necessary to cover the plants for a prolonged time with bags, and to atomize them

20 frequently with water to secure infection. A considerable decree of infection wa3 secured in this way, vet it is probable that this

Septoria does not occur commonly on L. can adensis in nature. The disease characters on L. canad ensis were different from these seen upon L. sat iva or L. 3ca riola as the spots were smaller, had no ten- dency to spread, were darker in color, and commonly had a vellow zone for a border. Thev were angular in outline, usually 1 mm, to

5 mm. broad, and the central area wa3 brown to black. Pvcnidia and spores were present.

Fspeciallv favorable conditions were also provided in order

to infect Sonchus a sper , still the transfer of the fundus to this more distantly related host was net difficult for more than 25^- of the inoculated leaves were irore or less diseased. Such infection raises a question of identity between Septoria lactucae and certain of the species of Septoria described upon the genus Sonchus. An examination of exsiccati showed a close similarity between S. lactucae and S. sonchif olia, but mere extensive dataare needed to prove their identity. The spots formed upon Son chus asp_er by S. lactuca.e were gravish-brown to black, ang-ular in outline, 1 mm. to 5 mm. in width, and contained reproductive bodies.

If Septoria lactucae is compared with S. lactucicola, it is found that the two species are different in their adaptation to their hcst3 as well as being different morphologically, although the ho3t ranges of the two almost coincide. Fach infects Lac tuca

3at iva , L. 3cariola, L. canadensis , and can pass over to the ~enus

Sonchus. Yet S. lactucae is best adapted to L. sat iva and L. scar-

iola, grows ]ess readily upon L. canad ensi 3, and infects S. asper

somewhat, while S. lactuci cola is best adapted to L. canadensis

grows less readily upon L. sativa and L. scariola . and infects S. oleraceus slightly.

Diagram III. Infections with Septoria lact icae from Lactuca sat iva and Lactuca scariola.

Table III 32

Infections with Septoria lactucae Pass,

from Lactuca sativa and Lactuca scariola

Condi- No. of

Date tions i Plants inoculated leaves infected & No. of Original host Lactuca sativa inoculat ed spots Remarks

Julyll f-b Lactuca scariola 10/10 many Spots like those on this host in nature

Sept. g-b 3 Lactuca canadensis 0/4 Feb. 10 g-b 3 ii it 0/4 0/8

Sept. g-b 3 Sonchus oleraceus 0/6 few Small spots but Feb. 10 g-b 3 n n 0/7 no pycnidia 07T3

Sept. g-b 2 Taraxacum officinale 0/6

Original host Lactuca scariola

Julyll f-b Lactuca sativa 5/12 10 Spots like those N It Mar. 2 g-b 3 ±91*9. 75 on this host in 15/22 85 nature

Julyll g-b 4 Lactuca canadensis 0/5 Spot 8 brown to Mar. 2 g-b 7 n it 8/A2 53 black and rela- 8/17 tively small

Julyl2 g-b 3 Sonchus asper 0/8 Nov, 11 g-b 3 « it 3/6 Pycnidia and Mar. 2 g-b 5 5/30 spores 8/3*4

July20 f-b Prenanthes sp. 0/6 Nov. 11 g-b 4 it it 0/8 0/14

Nov. 11 g-b 4 Sonchus uiiginosus 0/5

Nov. 11 g-b 4 Taraxacum officinale 0/7

VII Sept r i a tritici Desm.

There was some uncertainty regarding the name of the species of oeptoria from wheat used, in these experiments. In the literature two species are frequently mentioned as associated on

Tri ticum vulgare ; viz, Septo ria t ritic i Desm., described in 1842,

and S. grami num Desm. , described in 1843. To the letter is usually ascribed the damage done to winter wheat and other cereals by Septoiia. Cavara (4, p. 21) in studying the forms of Septoria on wheat was confronted by so much disagreement that he revised the descriptions of Desmazieres as follows:

" Die Septori a tritici Desm. (PI. Cryntog. no. 669), welche anfangs gelbe, dann rostbraune, und endlich weissliche

Flecke durch die ZerstSrung des Parenchyms bildet , hat gew5hn-

lich gef&cherte Sporen von 50 bis 60 X 1.5-2>^ ; sie sind von faden- fo'rmiger Gestalt mit hffuf ig ein wenig aufgetriebener Mitte. Die

Scheidewande sind von Desmazieres nicht bemerkt worden.- Die

Septoria grai^inujn Desm. (PI. Cryptog. no. 728) besitzt Perithecien die dem blossen Auge nicht wharnehmbar und kleiner und dichter gestellt als bei der vorigen Art sind; sie bilden durch ihre Ver- einigen la'ngliche grau nebelige Flecke. Die Sporen sind etwas feiner als bei Septoria tritici und ein Ende ist dicker als das

andere; sie raessen 40-50 X 1-1. 5>> , sind nicht gefachert, zeigen aber raehrere TrSpfchen."

Upon the basis of this revision Cavara compared all the exsiccati specimens available to him. His comparisons led him to make this statement: (4, p. 22) " V/enn men sich nun Rechenschaft

24 von der grosser! Veranderlichkeit der Charaktereigenschaften giebt, wie Form una Far be der Fie eke, Grdsse der Perithecien, Vorhanden- sein or Fehlen der Scheidewande bei der Gattung Septori*, so wird

es sehr welirscheinlich das S. graminum und tritici s nur

Formen einer einzigen mycologiuchen Art sind und die sich ergeb- enden Differezen vielleicht nur der Verschiedenartigkeit der Wirts pflanze zuzuschreiben sina"

The collections of Septoria. from wheat used in the present investigation were made from one plot of Turkey Red winter wheat at the Illinois Experiment Station. In June, at a time when the grain had fully headed, nearly all the leaf blaaes were more or less infected. After harvest volunteer clumps of whent come up from heads scattered on the ground, and in (October the leaves in these clumps were badly attacked by Septoria. Collections for purposes of inoculation were made at this time, and also in .Decem- ber and January. After the ground froze the clumps were cut out, allowed to thaw slowly, and v/ere then kept in a closed collecting can in the greenhouse. In a few days the pyenidia were forming spores abundantly.

The spores from the above source, collected either in summer or winter, were septate, and the pyenidia were brownish-black, round to elliptical when viewed from above, usually arranged in row3 between the leaf veins, and were often visible to the naked eye. The ipores from the leaves of the plants that haa been frozen and then treated as mentioned were much longer but somewhat more slender than spores from material collected in other ways.

Septntion was difficult to detect, and seemed to be absent in some cases, but proper staining made it apparent in the most of the

25 Table IV

Infections with Septoria tritici Desm.

from Trit icum vulgare

No. of leaves Condi- infected & Date tions Plants inoculated inoculated Pemarks

Oct. 23 g-b 3 Triticuir; vulgare 45/53 Check Jan. 19 (var. turkey red) 30/55 n 75/108

Oct. 23 g-b 3 Trit icum vulgare 46/48 (var. home grown spring)

Dec. 28 g-b 4 Triticum dicoccum 0/20 Leaves blanched Jan. 19 g-b 4 n w 0/20 tt it 0/40

Jan. 19 g~b 4 Triticum compactum 0/10 Leaves blanched

Jan. 19 g-b 4 Triticum polonicum 0/12 Leaves blanched

Dec. 28 4 Triticum durum 0/20 Leaves blanched Jan. 19 g-b 4 It M 0/20 0/40

Oct 23 g-b 3 Secale cereale 0/25 Jan. 19 g-b 4 tt w 0/35 0/60

Oct. 23 g-b 3 Hordeum vulgare 0/27 Leaves blanched Jan. 19 g-b 4 n tt 0/30 0/57

Oct. 23 g-b 3 Avena sativa 0/28 Leaves blanched Jan. 19 g-b 4 w tt 0/30 « tt 0/58

Oct. 23 g-b 3 Aeropyrum repens 0/21 Jan. 19 g-b 4 tt tt 0/9 0/30

Oct. 23 g-b 3 Poa pratensis 0/17

Oct .23 g-b 3 Phleum pratense 0/30

Oct. 30 g-b Festuca e^atior 0/30

Oct. 23 g-b 3 Dactylis glomerata 0/15 Jan. 19 g-b 4 n tt 0/30

26 Table IV continued

Infections with Septoria tritici Lesm.

from Triticum yulgare

Oct. 23 g-b 3 Bromus secalinus 0/40

Oct .23 g-b 3 Loliura italicum 0/35

Jan. ly g-b 4 Setaria glauca 0/15

Oct. 23 g-b 3 Digitarie sanguinale 0/35

Susceptibility of varieties

Jan. 19 g-b 4 Turkey Red 30/55 Spots large with numerous pycnidia

H M g-b 4 Beloglina 26/50 As above

it » g-b 4 Pesterboden 23/54 Spot 8 intermediate in size, pycnidia. few

II M g-b 4 Mnlnkof 28/65 As above

H II g-b 4 Hungarian 8/32 Spots few and small with very few pycnidia

Diagram IV. Infections with Septoria tritici from Trit icum vulgare

2? spores. The pycnidia. under these same conditions were often more than double their ordinary size, but were unchanged in shape.

Infections obtained upon wheat seedlings by inoculation with these longer types of spores always resulted in pycnidia of normal proportions, and in shorter spores with more definite septa.typ- ical of the fungus found on the leaves of wheat in summer or fall,

A more comprehensive description of these morphological variations will be given in a subsequent section.

Whether one base his conclusions upon Cavara's revised descriptions, or upon his assertion that Septori a tritici a.nd S. gram inum are probably identical, the fungus at hand must be desig-

nated S. tritici Desm. , in one case on account of morphology, and in the other on account of priority. Yet circumstances indicate that the organism is the same as that, often reported as injurying

wheat and other cereals and called 6. jgrjsuninjim. ( 4 , 14 , 16 , iy ) . It is well to note, however, that all the inoculation experiments, the results of which are summarized in table IV, were made v/ith spores taicen from the leaves of volunteer wheat collected in late fall and winter.

Seedlings grown in flower pots were exclusively used for

inoculation with Septoria tritici . Following inoculation the bags were placed over the pots for three days, but were removed temporarily each aay to atomize the leaves with water. On the fifth to sixth day yellow spots began to appear on the leaves of wheat. These spots enlarged and coalesced until half or more than half of the bla.des were yellow and drooping. Unless the plants in the greenhouse were covered with the bags for a second period or set in a large moist chamber, the yellow blades would dry and

28 no pycnidia would develop. The first pycnidia were observed on the twelfth day, and were distributed in greenish "lalanus" of tissue, which fact showed that the yellowing of the entire leaf surface was not all directly uue to the attack of the Septoria.

That the Septoria of wheat under consideration is limited in its host lange to the varieties of Triticum vulgare appears conclusive from the results of the infection experiments given in table IV. In some instances, especially with the other species of Triticum, and barley, oats, and rye, there was bls.nching of the inoculated leaves, but this progressed uniformly backward from the leaf tips and was not aue to a parasite. This reaction was also subsequent to the appearance of the irregular yellow

spots on the leaves of Tri tic am vulgare . No pycnidia were ever detected upon any plant except upon those in which positive infection is recorded. It is interesting to note that the fungus could not infect hosts closely allied to Triticum vulgare such as

T. durum, T_. compact urn , and T_. d i c c c urn . If the forms of Sep- toria upon the species of Gramineae listed below are ell morphologically alixe, as there is reason to believe, it is evident that

Septoria trit ici consists of more than one, possibly several biologic forms. To determine the infection-powers of the form on each of these grasses presents a profitable field for further ex-

periment .

The frequent occurren ce of this Septoria on the dead or over-wintering leaves of wheat suggests that it may be more saprophytic than parasitic in its habits, but the narrow limits of host range indicates a parasitic nature. If it were saprophytic it should have shown a disposition to attack the leaves of the other

29 cereals under the warm and moist conditions provided in the experiments, especially after the plants had become weakened.

Hosts reported for Hosts reported for oeptoria graraJL nura Septor ia. trit ici

Triticura vulgare Triticum vulgare Hordeum vulfeoxe Triticura caninum Avena sativa Glyceria fluitans Secale cereale Brachypodium Poa annua Pestuca Poa corapressa Poa pratensis Broraus sterilis Celamagrostis epigeios Calamagrostis langsdorfii Digitaria sanguinale Panicum scribnerianum Paspalum orbiculare Alopecarus agrestis Avena. planiculmis Car ex riparia

Two series of inoculations were conducted to ascertain whether there was any variation in susceptibility among the vari-

eties of Tritic ura vulgare * In the first series Turkey Red,

Minnesota Reliable, Red Cross, Red Hussar, and Home Grown Spring were about equally infected. Pesterbouen and Malakoff were infected to a less degree than the above varieties, while Hungarian was but slightly attacked. In the second series the more detailed data at the end of table IV were recorded, in terms of percentage

Turkey Red and Beloglina each gave about 52 % infection if the leaves that had one or more spots of infection were counted. Pes- terboden and Malakoff gave 43 % and Hungarian only 25 %. This is not a correct basis of comparison, however, for the individual leaves of Turkey Red and Beloglina had the largest and moat numerous spots with pycnioia in the greatest number. In Hungarian often but two or three pycnidia were present on a leaf, while with

Pesterboden and Malakoff the case was intermediate.

VIII Sejpjfcojfia malvic ola E. 4 M,

A Septoria determined as Septoria mal viccla E. & H. was collected upon \!alva rotund folia near Hutchinson, Minn. The common mallow is the only host reported for this fungus. The fact that this Septoria crave almost lOOf infection (Table V) when transferred to Althaea rosea, the hollyhock, suggests that the fungus may be identical with Sep toria. fai rmani E. & M. described on this plant. The identity appears evident when the spot characters produced upon the hollyhock by the mallow Septoria are compared with those of S. fai rmani as described and as found in exsiceati. The

spots on the mallow are smaller and a.re commonly surrounded by a broad yellow zone. When the fungus was transferred to the hollyhock, this yellow zone did net appear, but instead only the narrow black border, coincident with the limiting lea.f veins was present,

features given for S. fai rmani. When these spots obtained by the

inoculation of the hollyhock were ccrrpared with S. fai rmani. "North

American Fungi" No. 3557, there was agreement in all essential points. In the morphology of the funeri the specimens shewed no dis-

tinctions beyond the limits of variation, and the morphological dif-

ferences given in the literature are within the ranee of personal

error. The ease with which the fungus passed from the mallow to the

hollyhock indicates that this may happen in nature. The plants in-

oculated in the open were somewhat shaded, and were covered three

da.ys with a bell jar. The evidence is sufficient tc prove the iden-

tity of S_. rral vicola and S_. fai rmani. Leaves of both Althaea rosea

and r/'alva. rotundifolia infected by inoculation with 3pores from the

latter host are illustrated in Plate I, Figures 5 and 6.

Tab "1 ^ V

Infections with Septoria malvicola E. & M. from Malva rotundifolia

No. of leaves Condi- infected & Date tions Plants inoculated inoculat ed Pemarks

Aug. ej 1* f-b 3 Malva rotundifolia 9/12 i uc'ut. j H! siiiy 53 >-< u v a

n n f-b 3 Althaea rosea 8/8 Spots like those of the Septoria commonlv seen on the hollyhock M ft f-b Abutilon theophrasti 0/6 ti n tt ti g-b 4 0/4 0/10

it n f-b 4 Hibiscus syriacus 0/7

Mar. 2 g-b 4 Malva rotundifolia 10/15 Oldest leaves first and most

11 Sav XAy CO uCU

n rt g-b 4 Althaea rosea 6/7

Table VI

Infections with Septoria scrophulariae Pk.

from Scrophularia marilandica

Feb. 12 g-b 3 Scrophularia mariland- 5/5 Check, many spots ica in 10 davs

g_b 4 Verba8cum thapsus 3/5 Twenty spots in 18 days, no pycnidia

ti tt g-b 4 Verbascum blattaria 0/10 A few spots in three weeks, probably due to S. verbascicola

« n g-b 4 Antirrhinum majus 0/10

tt n g-b 4 Verbena urticifolia 0/10 Some minute brown spot9

n n g-b 4 Verbena hastata 0/10 Rome minute brown spots

n r g-b 4 Linaria vulgaris 0/20

32 IX Septoria scrophulariae Peck

Verbascum t hap sua was the only host outside the genus Scro- phularia to which Septoria scrophulariae would transfer, and in this case spots without pycnidia. developed. These spots are believed to represent a real cross-infection since the spores applied were from a pure culture, and the disease was confined to the area inoculated on each leaf. There is no proof that this Septoria does not

also attack V. blattaria . but if this host is susceptible, the degree of susceptibility is very slight. The plants of V. blattaria transplanted from the field could not be completely freed of S. verbas -

cicola , and this parasite continued to appear to vitiate the results of certain experiments. Whether the delayed development of a few

Septoria spots on V. blattaria was due to S. verbascicola from this extraneous source or to spores of S. scrophulariae applied in inoculation could not be easily determined.

A resemblance in many respects between S* scrophulariae and

S. verbascicola will be discussed in a following section.

Diagram V. Infections with oeptoria scrophulariae from Scrophularia maxilandica

X Septoria cgnvolyuli Desm.

Septori a sejotulata sp. no v.

In the Saccardian description of Sep toria convolyuli Desm. both Convolvulu s arvensis, and C. s_ep_ium are given as hosts, hut the following- distinctions in morphology are noted in the forrrs from the respective hosts: M In forma Conv. arvensis observavi sporulas

35-50 x 1-1. 5, aciculares minutissime 5-6 guttulatas v. septulatas hyalinas; in forma Calystegiae sporulas 35-40 x 1-1.5, continuas, hyalinas utrinque obtusiuscuias. An differentiae constantea "?

(Saccardo 3, p 536) ,

The forrr3 of Septoria collected at Urbana upon C. arvensis and C. sepium fitted respectively the characterizations of the above forms as quoted. One hundred spores from C. arvensis had an average length of 44 microns, whereas a like number of spores from

C. sepium averaged only 35. 5 microns. There was a difference in average spore length whatever the conditions under which the two forms were compared. The distinction mentioned regarding: spore tips is not evident, but the spores from C. a rven sis were always the more definitelv septate. With respect to host characters there is little by which the forms can be distinguished.

The results of the infection experiments (tables VTI and VIII) proved that the forrr.3 of Septoria from these two bindweeds are likewise distinct in their powers of infections, for each can cause vigorous infection upon only its original host. The fungus from C. arv nsis does infect C. sepium to some degree, with 9 out of 39 leaves inoculated showing a few small disease spots, but this is to be compared with 39 out of 45 leaves upon the original host, on which the spots were ordinarily so numerous and so spreading that

34 TABLE VII

Infections with Septoria septulata sp. nov. from Convolvulus arvensis

^o. leaves

Condi-- infected * 1 Mo. of Cat e t ion9 Plants inoculated inoculated 3P0t S

Oct. 22 g-b 3 Convolvulus arvensis 15/20 rany 1 ^Jheck, most of « it Nov. 14 sr-b 3 ii 4/5 the leaves « ti Mar. 2 g-"b 3 It 20/20 tilled OS/ 45

June 26 g-b Convolvulus sepium 3/20 11 Pycnidia and Oct. 22 g-b 3 « ti 12 3pcres but spots Nov. 14 g-c o it ti irery small Mar. IS g-b 4 II M 1/9 1 9/39 29

Nov. 14 g-b 3 Ipomoea rurpurea u/ o Mar. 2 g-b 3 ii n 0/10

Oct. 22 g-b 3 Iporr.oea batatas 0/10 Nov. 14 g-b 3 0/7

liar. 2 g-b 4 Ipomoea lesrii 0/10

-i:ar.II o -n &9 g-b 4 Iporr.oea setosa 0/4

TAFLTT VIII

Infections with Peptoria convolvuli D e sm from Convolvulus sepium

Sept. 10 f-b 4 Convolvulus sepium 6/7 31 Check, spots Oct. 10 g-b 4 n n 5/5 12 large Nov. 14 g-b 4 ti H 5/6 16 Mar. 18 g-b 4 M l» _4 If/ 20 bo

Sept. 10 f-b 4 Convolvulus arvensis 0/7 Few minute spott; Sept. 25 f-b 4 « « 0/12 no pycnidia, in- Nov. 14 g-b 4 « ii 0/25 fection doubt- Mar. 18 g-b 3 « n 0/35 ful 0/79

Nov. 14 g-b 4 Ipomoea purpurea 0/5 liar. 18 g-b 4 0/5 0/10

Oct. 22 g-b 4 Ipomoea batatas 0/5 Nov. 14 g-b 4 n ti 0/10 0/15 Mar. 18 g-b 4 Ipomoea learii 0/20 0/5 ..ar. io K-b 4 Iporr.oea setC3a

35 the leaves were entirely killed. When C. arvensis was inoculated with the fungus from C. sepium no disease spots with pycnidia were obtained upon any of 79 leaves inoculated, although minute brown spots, usually less than 1 mm. broad, were often formed. The sig- nificance of these spots is unexplained, though they rray represent incipient infection.

Septoria convolvuli is reported upon I porno ea purpurea , but all attempts to infect species of Ipomoea were futile.

Were the two forms of Septoria under consideration to be accepted as belonging to a single species, here would be a well established case of biologic specialization. Put since the form3 have been proved to be different both morphologically and biologica.lly, they should each have specific rank. The name S. convolvuli is re- served for the form on C. sepium upon which host the type of S, convolvuli was described (6). For the form upon C. arvensis the following new species is proposed:

gejstcria septula ta sp. nov.

Spots orbicular, then irregular and confluent, light to dark

60-90/w brown; pycnidia mostly epiphyllous innate, globose, , pro- truding with a prominent ostiole, 20-3CUA/; spores curved or flexuous, one end narrow with a more acute tip, 3-5 septate, 30-50/w long by

l-2^wide. Habitat: old or fading leaves of C. arv ensis .

This species appears to be the equivalent of the form on _C. arvensi described in Saccardo'3 "Syiloge Fungorum", Vol.3, p 536.

5lo

Diagram. .VII. Infections with Septcria septulat a frorr Convol- vulus arv gngie and S. convolvull

from j_. sen ium

Diagram VIII. Infections with Sep toria verc asci col a frorr Verbaecum blatt ari a

XI Sep tori a verbascicola P.& C.

Septo ria verbascicola appears to be somewhat adaptive in its host relations. The infection of Scrophula ria irari landica was as vigorous as that on the original host when conditions were rrade favorable for the fungus. This was done by keeping the inoculated plant constantly covered with a bell jar for ten days, and by atom- izing the leaves daily with water following the application of the spores. At the end of this period under cover disease spots began to appear. The plant used was a small shoot which sprang from under- ground parts transplanted from a woodland in fall. It seemed to re- tain its full vigor notwithstanding the rather unusual conditions provided in the experiment. As may be seen from Table IX the infection of Scrophula ria marilandica secured without these special conditions was slight. When the fungus was transferred back to Verba scum

bla t taria , the original host, there was heavy infection, though the inoculated plant was inclosed no more than two days.

The disease spots formed upon Scrophularia marilandica were essentially like those upon Verb ascum blat taria. Upon either host the central area of the spots was ashen in color, while the border was reddish-brown to purplish. Under humid conditions a dull green zone appeared outside the reddish-brown ring, indicating the destruc- tion of new tissue by the encroaching fungus, or the entire area, of the spot would be dull green to black. This latter phenomenon was to be seen chiefly upon the original host. Pycnidia were numerous upon both hosts under the humid conditions.

Septoria scrophulariae Pk. , which is common upon Scrophularia marilandica in nature, forms spots scarcely dist ins-uishe.ble from

38 those above described. The irost important distinction is the scanty number of pycnidia; not more than a half dozen are often to be seen in a spot. The spores and pycnidia of S. verbascicola and S. scro- phuiariae frorr material collected near Urbana were so similar that it was suspected that the forms were identical. The fact that the former infected Scrophularia marilandic a readily made such identity probable, but this view was shown to be questionable when it was found that the latter infected V. blattaria slightly if at all. It was discovered, moreover, that the two fungi can be separated easily in culture, as S. verbascicola produced pycnidia and spores upon all media used, while S. scrophulariae seldom did; the growth of the first was chocolate colored, that of the second v uff.

Septoria y erbasciccla attacked V. thap sus readily, but few spore-bearing structures developed. The spots attained a size as

great as these upon V. blatt aria , were angular in outline, and purplish in appearance. It seems probable that this host is infected in the field, although no proof of this was secured. Leaves of mul- lein were collected with similar disease 3pots, some of which were very large but as no pycnidia were found, it cannot be stated that the injury was caused by Septoria.

The rather constant formation of small brownish s^ots upon the species of Verbena following inoculation with 8. y e rb as c i_c ola would appear to indicate incipient infection, yet the failure to obtain pycnidia, and to prove that the spots were not due to other causes leaves the matter in doubt.

Table IX

Infections with Septoria verbascicola B. & C.

from Ve rb s cum blattaria

No. leaves Condi infected & No. of Date tions Plants inoculated inoculated spots Remarks

Aug. 30 f-b 4 Verbascum blattaria 3/6 8 Check Oct. 30 g-b 3 it it 5/5 43 M Feb. 12 g-b 3 W II 10/10 75 It 18/21 126

Aug. 30 f-b 3 Verbascum thapsus 3/5 8 Pycnidia & spores n It Oct. 30 g-b 3 n i/6_ 10 It w 7/11 18

Oct. 30 g-b 3 Scrophularia mari- 2/8 5 Pycnidia & spores Jan. 15 g-b 10 " landica 5/5 33 n n ii 7/13 38

Oct. 30 g-b 3 Verbena urticifolia 0/7 f ew Infection doubtful Feb. 12 g-b 6 it ii 0/5 few n tt 0/12

Oct. 14 g-b 3 Verbena bastata 0/5 few No pycnidia, in- Feb. 12 g-b 5 It tl 0/5 few fection doubtful 0/10

Aug. 30 g-b 4 Verbena stricta 0/10

Mar. 15 g-b 3 Linaria cymbalaria 0/20

Aug. 30 f-b Linaria vulgaris 0/30

Aug. c w £— D C niiu irniiuuiii rnsLj us Jan. 15 g-b 3 it ti 0/10 Fee, 12 g-b 5 I tt 0/10 0/32

Mar. 15 g-b 4 Digitalis gloxini- 0/10 aef lora

XII Septoria cirsii Niessl.

Upon Ci rsium arvense . Septo ria cirsii is an active parasite, both in the field and when brought into the greenhouse. In the field it attacks chiefly the lower leaves, and causes large, usually irregular, brown, dry spots, mostly alone: the leaf margins, but often the whole leaf is involved. In the greenhouse growth is similar but mere rapid and extensive. Inoculated leaves are generally completely destroyed. The fungus is also reported upon Ci rsium dis -

color .

According to the infection experiments conducted Septoria cirsii can make a weak attack upon both Cirsium discolor and C.

lanceolatum . The spots formed upon these hosts resemble those upon

C. arvense , but are smaller and very few in number. The difference is the vizor of attack upon these hosts is also well demonstrated by the length of the incubation period and the rate of development.

Upon Cir9ium arvense spots appeared in thirteen davs after inoculation, pyenidia were observed in fifteen davs, and in four weeks the whAle plant was dead from the spread of the disease. Upon Cir- sium lanceolatum spots were first visible in twenty days, and in four weeks they haa attained a breadth ranging from 3-8 mm.

Pyenidia and spores were found at this time. The plant had to be kept in a moi3t atmosphere to bring about the formation of many spores. The incubation period and rate of development were similar

in Ci rsium discolor . There were no noticeable differences in the morphology of the fungus upon the various hosts.

The relation of maturity of the spores to the presence of septa was often noticed in this species. Where spores were made to _

41 Table X

Infections with Sep toria cirsii Niessl.

f rorr. Cirsium arvense

. i Nn of leaves Condi- infected & No. of Date tions Plants inoculated inoculated spots Remarks

moTiTf cj une&o f-b J- L X Lull ct x v cii ci c UfO nieiuy bnecK, is av ss kiii eu Oct. 24 g-b 5 ft tl 5/5 it n it it Feb. 12 g-b 4 n R 15/15 ti « it n 25/25

d «J un e<3 g-0 \j 1 T3 lum w/ o Oct. 24 g-b 3 It " turn 1/5 4 Pycnidia & spores « It II Feb. 12 g-b 5 3/8 8 R It 4/18 12

June28 g-b 3 Cirsium discolor 0/10 Oct. 24 g-b 5 « ii 2/10 oi Pycnidia & spores Feb. 12 g-b 5 n 2Zio 2/30

Aug. 30 f-b Cirsium hilli 0/5 Oct. 24 g-b 4 n n 0/15 Feb. 12 g-b 4 n « 0/10 0/30

Table XI

Infections with Septoria brunellae F. & H.

from Prunella vulgaris

Sept. 5 f-b Prunella vulgaris 3/8 8 Check Mar. 9 ©cr—h 4 11 It 5/15 11 8/23 19

Sept. 5 f-b Nepeta cataria 0/5 Mar. 9 g-b 4 0/10 O/Ib Sept. 5 f-b Nepeta hederacea 0/10 Plants shaded Sept. f-b Monarda fistulosa 0/10 Plants shaded

Sept. f-b Blephilia hirsuta 0/5 Plants shaded Mar. 9 g-b 5 0/10 07T5 Mar. 9 g-b 4 Salvia coccinea 0/10

develop abundantly in a moist chamber, septa could not be seen at

all, or could be seen onlv faintly after staining with iodine. In

old spots or where the ?rowth had been slow, the septa were plain-

ly seen. Spcres obtained from an old culture upon corn meal agar had 3epta that were strikingly definite. Plants inoculated with

spcres from this culture were heavilv infected.

Diagram IX. Infections Diagram X. Infecticn3 with Qeptoria with Septoria cirsii brunellae from Prunella vulgaris,

fron Ci sium arvense .

XI11 Septoria brunellae E. & M.

The inoculations with Sep toria brune llae gave negative results in all case 8 except upon the original host, which facts indicate that the fungus is probably limited to Pru nella vulgaris in its host range. Upon this one host it is a marked parasite; frequently the whole surface of some leaves is discolored by the confluent spot s.

XIV Septoria lycgpersicl Speg.

The inoculations with Sep toria lycopersi ci gave results in agreement with those published by Norton (22) who inoculated several plants related to the tomato in humid inclosures. He states that were spots developed better and spores/ larger on potato and Solanum carolinense than on tomato, while on Datura the spots were slow growing, light colored, and small-spored. In the present experiments the spots on potato were darker in color and smaller than those cn tomato, while those on Solanum carolinens e were likewise darker in color, but larger, and often coalesced until the leaves were destroyed. The spots on Datura were very similar to those described by Norton.

Spores developed upon the potato and Solanum carol in ense produced full infection of tomato leaves.

44 Table XII

Infections with Septo ria lycopersici Speg. from Tomato

No. of leaves Condi- infected & No. of Date tions Plants inoculated inoculated spots Remarks

Sept. 16 g-b 2 T o mat o 5/5 30 Ch^ck

it g-b 2 Sow 1X CL*1a tuttiLAiil "tutoproW U v C A ^ OsumLA ilJ O/5 Mi tit it; p Q"00+.fl possibly infection

tt it g-b 2 Solanum carolinense 0/5 Minute spots, possibly infection

tt It K-b 2 Solanum nigrum 0/5

tt tt g-b 2 Datura tatula 0/5

tt n 2 Pnysalis heterophyl- 0/6

R it g-b 2 PVivanT i a 1 anppfll at.a 0/8

Mar. 2 g-b 3 Tomato 10/13 many Check, leaves n 18 g-b 5 n 10/10 it killed 20/23

tt 2 g-b 5 Solanum tuberosum 10/10 many Pycnidia & spores

N 2 g-b 5 Solanum carolinense 4/4 many Pycnidia & spores It 18 g-b 5 tt it 6/8 n It It M 10/12

« 18 e:-b 5 Datura tatula 5/10 24 Few pycnidia and

18 g-b 5 Solanum nigrum 0/20 Few yellowish, possibly infection

N 18 g-b 7 Petunia 0/12

XV Septoria lep i diicola E. & M.

Septor ia lepidiicola is reported upon only Lepidiurr vlrglnlouBj

and L. apetalum . The data in table XIII ehow that the funerus on these two bests is identical biologically as well as morphologically.

The negative results obtained when plants belonging to other genera of Cruciferae were inoculated indicate that the fungus is confined tc the species of Lepidiurr. No data are at hand to show whether it will attack more than the two species of the genus. The failure to obtain signs of incipient infection, such as spots without spcre- bearine-bodi es, upon species outside the genus Lepidiurr. proves that the fungus has little adaptability. It may be possible, however, to obtain growth in some form upon the remaining species of Lepidium that have not been inoculated experimentally.

Diagram XII. Infections with Septoria lepidiicol a from Lepidium virginicum .

46 Table XIII

Infections with Septoria lepidiicola F. A V.

from Lepidium virginicum

No. of Check leaves inf ect ions Condi- infected & on original Date tions Plants inoculated inoculat ed host

June 6 g-b 3 Lepidium apetalum 5/12 7/20

Oct. 12 g-b 4 Thiaspi arvense 0/30 8/25 Oct. 23 g-b 4 it n 0/40 11/40 0/70 19/65

Mar. 6 B-b 3 Brassica alba 0/20 20/40

Mar. 6 g-b 3 Brassica arvensi3 0/10 20/40 June 6 g-b 3 n n 0/11 7/20 0/21 27/60

Mar. 6 g-b 3 Brassica juncea 0/20 20/40

Mar. 6 g-b 3 Raphanus sativus 0/5 20/40 June 6 3 ti n 0/7 7/20 0/12" 27/60

Mar. 6 g-b 3 Camelina sativa 0/20 20/40

Oct. 23 g-b 4 Sisymbrium altissimun 0/33 11/40

Mar. 6 g-b 3 Sisymbrium officinale 0/15 20/40 June 6 g-b 3 n tt 0/10 7/20 Oct. 23 g-b 4 n it 0/2C 11/40 0/45 38/100

June 6 g-b 3 Capsella bursa-pas- 0/7 7/20 Oct. 23 g-b 3 n " toris 0/20 11/40 0/27 18/60

XVI Septoria hel ianthi F11.& Kell.

The following plants have been reported as hosts for Septoria helianthi

1. Helianthus grosseserratus 5. Helianthus petiolaris M 2. " annuus 6. lenticularis 3. " doronicoides 7. " strumosus 4. " californicus 8. Heliopsis laevis

In the vicinity of Urbana this Septoria was a common parasite of H. gro sseserratus, upon the lower or shaded leaves of which it causes grayish-black spots that often become more th&n a centimeter in

breadth. It was less commonly found upon H. tuberosus . upon which the spots are smaller, and rarely upon H. annuu3 and H. rigidus.

In three separate trials H. grosseserratus was readily infected with its own form of Septoria without maintaining a humid atmosphere bevond three days. The other species of Helianthus infected by the Septoria from H. gros seserratus are shown in diagram XIII.

Except for H. tuberosus, which grew natively in the greenhouse vard, all these cross-infections were upon seedlings grown in the greenhouse. Proof of infection was obtained only by detaching spotted leaves and laying them in a moist chamber to induce the development of pycnidia. For this reason, as well as on account of the attack of insect pests, no accurate data could be secured concerning the degree of susceptibility of certain of the host9, consequently some of the records are wanting. It is clear, however, that the original host is the only one that is congenial for this Septoria from H. g rosseserratus. for on its own host alone spore-bearing bodies devel oped naturally under the ordinary greenhouse environment. The infec tion of small shoots of H. tuberosus covered, out of doors, with a bell jar was relatively feeble. Numerous spots developed but the

48 author could not satisfy himself that they were entirely due to Sep- toria, on account of t'r e scanty appearance of pycnidia, and because a Phyllost icta, appearently saprophytic however, often appeared when the leaves were placed in a moist chamber.

It appears significant that the Septoria from H. rigidus did not infect H. grosseserratus when the plant, inoculated in the field was bagged five days, and that the fom~ from H. tuberosus did not give ready infection of H. grosseserratus in the greenhouse. The failure of infection in the series of inoculations made upon mature plants of various species of R'elianthus on Sept. 16 in the greenhouse

(Table XIV), and also upon similar host3 in the field in summer, in which spores from H. gro sseserratus were used, may have been due to the age of the leaves, yet the checks on the original host gave positive results. A corresponding set of hosts were inoculated at these same periods with spores from H. tube rosus with negative results, but these were not recorded in the table as no checks were used. Still in these cases the spores germinated well in laboratory tests.

All the data obtained respecting SepJ;oria hel i ant hi indicate that the forms used in inoculation are not vigorously parasitic except upon their original hosts. These data are scarcely ample for definite conclusions, yet the results are in harmony with what was observed regarding the host range of this Septoria in the vicinitv of Urbana. If there are no fixed biologic forms, nevertheless there appears to he a degree of specialization upon the different species of Helianthus, possibly of a temporary nature such as Vontemartini

(18) claims for rusts.

49 Table XIV

Infections with Septoria helianthi Ell. & Kail. From Helianthus grcsseserratus ===^===- No. of leaves Condi- infected & Date tions Plants inoculated inoculated Ferr.arks

Sept. 16 e-b 3 Helianthus grcsseserra- 5/5 Check, leaves mature tus 32 spots

n n g-b 3 Helianthus occidentalis 0/6 Leaves mature

it H g-b 3 Helianthus rigidus 0/5 H tt

II tt g-b 3 Helianthus mollis 0/5 It It

tt it g-b 3 Helianthus tuberosus 0/5 II It

it tl g-b 3 Helianthus annuus 3/8 8eedlings,pycnidia in moist chamber 10 spots

it n g-b 3 Silphium int eg ri folium 0/5

n it g-b 3 Bidens cernua 0/20

July 5 e-b 3 Helianthus grosseserra- 7/7 Check, leaves young tus 26 spots

n it f-b 4 Helianthus tuberosus 4/20 Many spots, but few with pycnidia

n f-b 4 Helianthus mollis 0/7 Leaves young

n f-b 4 Helianthus rigidus 0/10 tt n

n n f-b 4 Silphium int egrifolium 0/5

Jan. 5 g-b 4 Helianthus rigidus plU8 Seedlings, only spot 3 formed

n n g-b 4 Helianthus argyrophyllus 4/20 Seedlings, pycnidia in moist chamber

tt it g-b 4 Helianthus cucumerifol- 6/20 As above

Xi nLa aO

Mar, 2 g-b 4 Helianthus grosseserra- plus Check, seedlings tus

it g-b 4 Helianthus californicus plus Seedlings, pycnidis in moist chamber

. a

Tabic XIV continued

Infections with Septoria heliant hi Ell. & Kell.

Har. 2 g-b 4 Helianthus annuus plus Seedlings ,pycnidi nanus flore pleno in moist chamber

it it g-b 4 Helianthus annuus plus A 8 above double primrose queen

n ii g-b 4 Helianthus rnaxirr.il ianus 0/10

it H g-b 4 Coreopsis lanceolata 0/20

n tt g-b 4 Rudbeckia laciniata 0/10

Spores f roir. Helianthus tuberosus

July 24 f-b 4 Helianthus tuberosus 4/10 Check, pycnidia and spores

N it g-b 4 H. grosseser ratus 0/5

ft it f-b 4 Helianthus rieridus 0/10

n it f-b 4 Helianthus mollis 0/10

it it f-b 4 Helianthus a.nnuus 0/5

it it f-b 4 ?ilphium int egrifolium 0/5

Spor es f rom Helianthus rigidus

Sept. 18 f-b 5 H. grosseserratus 0/8 No check, but spore germinated well in n n f-b 5 Helianthus tuberosus 0/5 laboratory tests

Diagram XIII. Infections with Septoria helianthi from Helianthus groase - aerratua

XVII Septoria. rubi lest.

Saptoria rubi was collected frequently at Urbana upon Rubus occidentalis, the black raspberry, but was never found upon any species of Rubus belonging to the blackberry, or Eubatus, section of the a'enus. Since plants of the blackberry were often growing in close enough proximity to R. occ identalis to enable therr: to become inoculeted with the Septoria, it appears that the species of blackberry were not susceptible to the fungus on the clack raspberry. During July thirty leaves each of P. occidentalis and a species of blackberry were inoculated with spores of Septoria from the former host. Both of these plants inoculated were young, and were growing- close together in a shaded place. In two weeks all the thirty leaves of R. occ identalis were thickly spotted with Septoria, but no trace of infection could be found upon the blackberry.

No definite conclusions can be drawn from the above observations, and the accompanying experiment. The facts may indicate that Septoria rubi is split into biologic forms, or that the species of blackberry in this locality are permanently resistant to the

Septoria. This fungus is reported, however, upon twenty-ei^ht or more species of Rubus, distributed among all sections of the erenus, a fact that makes it appear probable that all members of the genus are parasitized. If this be true the brief data above indicate the existence of biologic forms in S. rubi. The la.r

in species the cenus Pubu.3, of which so many are known to >^e hosts of Septoria, .Takes them an attractive field for a further study of biologic specialization.

XVI 11 Septoria at ro -purpu rea Peck

The presence of disease spots without pycnidia is the only

indication that Septoria at ro - purpurea from Aster cordifolius can

infect A_. ericoides and A. laevis . This infection wa3 obtained under prolonged humid conditions, and it is irrprobable that it oc-

curs often in nature. These asters have not been reported hitherto

as hosts for the fungus. This parasite, according to specimens in

the Herbarium of the United States Department of Agriculture, at-

tacks S olidago lat if olia and M acha e ran t h e ra aspera. The data indi-

cate that this Septoria is able to adapt itself to related hosts

to a considerable degree.

The disease spots on A. cordifolius are nearly orbicular, with

a central area of reddish-brown and a. margin of light green. In the

larger and older spots a gray area appears within the reddish-brown.

Upon A. ericoides the spots were a bright reddish color, with a mar-

gin of yellow, and a ragged and indefinite outline. The spots on A.

laevis were very small arc! brown, indicating that the fungus wa3

less well a.dapted to this host than to A. ericoides . As the spores

for inoculation were from pure culture, and adjacent leaves not in-

oculated shewed no symptoms of disease, it appears certain that the

spots in question were due to the Septoria applied.

Table XV Infections with Septoria atro-purpurea Peck from Aster cordifolius

No . leaves Condi- infected & No. of Date t ions Plants inoculated inoculated Spot 3 Remarks

Feb. 25 g-b 6 Aster cordifolius 11/18 46 Check

H It g-b 6 Aster ericoides 4/10 13 No pycnidia

ii n g-b 6 Aster laevis 4/15 9 No pycnidia

53 XTX General Discussion

Age incidenc e. - In the field Septcria. spots are more frequently observed upon the old or fading leaves of plants, but this is often due to the fact that the young leaves have recentlv expand- ed, and have not had time to become infected. The immature foliage is sometimes the mere susceptible, but this varies with hosts. From the results of the infection experiments the leaves of Lep idium virginicum appear to be equally susceptible at all ages. Old leaves of Lactuca scaricla in which the edges are cracked and drying are infected without difficulty, but the spots remain small. Partially grown leaves of this host are more ea.sily infected, and the spots becon e larger. The older leaves of Convolvulus arvensis and C. sepium are the m.03t heavily attacked, while inoculation of the immature leaves seldom produces infection.

The influence of age upon susceptibility is very well shewn

in Malva rotundif oli a and Althaea rosea . Disease first develops on the oldest leaves, in these of intermediate asre the incubation period has greater length, while those partially grown are resistant. P plant of Althaea res 3a with seven leaves was thoroughly wetted with

a suspension of spores of Septoria malvicola , and kept under a bell jar four days. In ten days the five mature leaves were spotted, the oldest having the heaviest infection. It was nearly three weeks before the sixth leaf shewed disease, and the seventh or youngest leaf inoculated continued healthy. In two trials with the mallow there were similar results, which are in accord with what one sees in the field, for there only old shaded leaves are badly attacked.

Suscept ibility of aif f erent leaf surfaces . - Equal areas of

54 the upper and lower surfaces of separate sets of leaves of a number of plants were inoculated, and other factors made as comparable as possible. In Lactuca scariola the upper surface gave the heavier infection. In Polygonum persicaria 39 spots, 1 mm. to 5 mm. in diameter, resulted on the top side against six spots, 1 mm. to 3 mm. in diameter, on the bottom. The leaf surface of Frigeron annuus in 150 separate areas, 75 above and 75 below, was inoculated with loop-

drops of a suspension of spores of Septoria erigerontis . and 14 infections from above and six from below were obtained. In contrast to these experiments the inoculation of the lower surface of leaves of Solanum carolinense with S. lycopersici gave abundant infection, while leaves inoculated above remained free of disease. No explanation of the above results can now be given, but the number of sto- mata, the character of the cuticle, the ease with which the suspension makes contact with the surface, and the light exposure may be factors.

Effec t of the mass of inoculum . - The effect of varying the concentration of the spore suspension was tested. One hundred areas upon F rigeron annuus were inoculated with loop-drops containing ap- proximately 100 spores, and an equal number of areas with loop-drops averaging 1 to 3 spores. In the former case there were 35 areas infected, and in the latter 15, but in the 35 areas over 80 points of infection were noted, while in the 15 areas only 18 were present.

Infections were obtained upon Lactuca 3cario la with loop-drops containing only 1 to 3 spores in 11$> of the inoculations.

Effect of wounding . - Certain leaves of young potted plants of

Malva rotundif olia were perforated with a fine needle in numerous places. All the leaves of each plant were inoculated over the entire

55 upper surface, and bagged for four davs. Only the perforated leaves became infected. Either the wounding had the effect of overcoming the resistance shown to be present in the youns- foliage, or the pierced epidermis afforded an easy entrance for the gerrr tubes of the spores.

Variations in the mo rphology of the fungus . - The more detailec studies of morphological variations were made with Septoria t ritici

and S. verbascicola , and the chief results obtained are shown in graph I by curves which represent the ranges of spore length in these fungi under a number of different conditions. No investigation was made of the factors causing these variations, but apparently they are due to differences in humidity. An increase in spore length when infected leaves were kept in a moist chamber for a period wa3 apparent in many other species of Septoria.

There was no alteration in spore length when S. verbascicola was transferred from Verbascum blattaria to Scrophularia marilandica under comparable conditions, as E and F of graph I 3how. Norton (22) reports that the spores of the tomato Septoria become longer upon

the potato and Solanum carolinense . but shorter upon Datura tatula than they were upon the original host, when the inoculations were made within humid inclosures. It is not stated whether he took account of environmental relations other than the change of host, but the lesser spore length upon Datura tatula makes it appear that the host has a definite effect on morphology. Still in the numerous crosses recorded in the tables of the present paper, no changes of spore length were observed for which environmental factors such as humidity might not account, at least in all cases in which the fungus and h03t were in compatible relations.

If all the species of Septoria have a corresponding relation to the environment, and if all have a broad range between the maxi-

in S. mum and minimum spore length, as trit ici and S. verbasoicola , it is obvious that many of the measurements now given in specific descriptions are far from accurate. The range of IS to 62 microns reported for S. sisymbrii is not of greater width than that shown for the fungi in graph I, and indicates care in measurement. It is not proper to compare dry herbarium specimens with material freshly collected, especially if it has been allowed to form spores in a moist chamber. It is well to make a record of the source of the material collected, and the conditions under which it was found,

cf. , graph I, C, D, G, and H.

Host limitations . - The experimental results thus far obtained indicate that the species of Septoria do net have a broad host range. Fach can infect vigorously one or a few closelv allied plants, and can infect to a less degree a number of hosts that stand in rather immediate relation to the vigorously infected ones. In

BOme cases this host range doe3 not extend beyond the limits of a genus, and In other cases includes but two or three related genera.

Although some of the fungi studied have been found to have approxi- mately the host range previously reported for them, in many instances the host ranges established by the experiments have been much more narrow than the reports on h03ts would lead one to conclude, a few forms being limited to the species upon which they are collected.

Notable illustrations cf this are the forms of Septoria on wheat,

- Convolvulus sep ium , and possibly upon Rubus occidental is and Hel ian

thus sp . Further infection experiments would doubtless reveal more cases of identity both morphologically and biologically among

57 forms now classed as separate species, while at the same time some of the present species would be shown to consist of more than one morphological form, or species.

The value of disease characters . - The variable nature of disease characters, as manifested by the host, has been well demonstrated. These variations are dependent upon the species, the age, amd the part of the host as well as upon environmental conditions.

On this account these characters lose much cf their value in taxonomy, but inasmuch as the host ranges of the species of Septoria are not broad, and the number of forms parasitizing a single host are very few, such characters may be of some use in distinguishing the parasites on individual hosts. For these same reasons the host itself will continue to be a valuable key in the determination of the fungus.

Biologic specializ ation ." The experiments herein described have not been broad enough to include fcrrrs from all the hosts reported for any species, especially with such fungi as Septoria rubi upon numerous members of Fubus, S. graminum or S. t rit ici upon several different genera of Gramineae, and S. polygonorum on various species of Polygonum. Great difficulty will inevitably be met in bringing together even a major portion of the respective forms for comparative study, such as would be necessary to firmly estab- lish the existence of biologic specialization, and to ascertain the number of biologic forms. Still such data as the present investigations furnish indicate clearly that biologic specialization ex- ists in many species of Septoria. This is shown by the fact that in many instances the Septoria from one host either fails to infect, or infects to only a slight degree, certain hosts upon whic v

58 morphologically similar forrs of Septoria qre known. In illustra- tion of this, the species of Septoria from wheat, fror pubu s occi -

dentqlis , and f ror certain species of Helianthus and Polygonum furnish examples.

XX Conclusions

1. The results of the present investigations indicate that certain species of Septoria are differentiated into biologic forms.

2. Although some forr-s show a degree of adaptability in host relations, in general the species studied are limited to one or to a few closely related hosts which they can vigorously infect.

3. In soiiie cases the host range does not extend bevond the limits of a genus, while in other cases two or three related srenera are included.

4. In many cases the host ransres established v v the experiments have been more narrow than the host indices indicate.

5. Disease characters, as manifested by the host, vary with the host and with environmental conditions, and arc therefore un- reliable in taxonomy.

6. Certain species of Septoria have been shown to vary con- siderably in morphological characters under different environmental conditions, and hence the value of measurements now gives in specific descriptions i3 questionable.

7. Inoculation experiments shew that Septo ria malvicola F. &

X.. and S. fai rrra.nl F. & F. are identical.

8. Similar experiments show that the form of S. convol vuli

Eesm. described upon Conv olvulus arvensis is biologically as well as morphologically distinct from the type form of S. convolvuli

described upon C. sepium . and is entitled to specific rank.

Literature Cited

1. Arthur, J. C. Cultures of Ureaineae. Journ. Myc. lj.:5Q-67. 190b.

2. Carleton ,1.1. A. Cereal Rusts in the United States. U.S. Dept. Agr. Div. Veg. Phys. and Path. Bull. 16. 1999.

3. • Investigations of Iiusts. U.o. Dept. Agr. Bur. Plant Indus. Bull. 63. 1904.

4. Cavara,J. Uber einige parasitische Pilze auf dera Getreide. Zeitschr. f. Pflanzenkr. 3:16-26. 1693.

b. iJieuicke , H. Uber den Zusammenhang zwischen Pleospora und Ilel- minthosporium-Arten. Centralbl. f. Bakt. Abt. 2. 9:317-329. 1902.

6. Desraazieres , J .B. Neuvieme notice sur quelques plantes crypto- games, la plupart ine'dites, re'cemment de'couvertes en Pra-nce, et qui vont paraitre en nature dans la col- lection publie'e par l'auteur. Annales des Sciences Naturelles. Ser. 2. 17:91-118. 1342.

7. Elliott, J. A. Taxonomic Characters of the Genera Alternaria and Macro sporium. Araer. Journ. Bot. jl: 439-476. 1917.

8. Eriksson, J. Ueber die Spezialisierung des Parasitismus bei den

Getreiderost Qilzen. Ber. Deutsch. Bot. Ges. 12 ; 292- 331. 1894.

z- 9. . Ueber die opezialisierung des Getreideschwart rostes in Sweden una anderen .rindern. Centralbl. f. Bakt. Abt. 2. 9_:654-6bS. 1902.

10. Gaxman,P. The Genus Septoria. Presented in Tabulation with Discussion. Unpublished.

11. Gilbert ,B.M, Biologic Forms of Black Knot. Phytopath. 3:246- 247. 1913

12. Hesler,L.K. Biological Strains of Sphaeropsis malorum. Phy- topath. 4:4b. 1914.

13. Hitchcock, A. S. and Carleton ,L.A, Second Peport on the .Busts of Grain. Kan. Agr. Exp. St a. Bull. 46. 1694.

14. Johnson.A.G. Septoria on Barley. Abstract in Phytopath. 6:117. 1916.

15. Levin, E. The Leaf-Spot Disease of Tomato. Mich. Agr. Coll. Techn. Bull. 2b. 1916.

16. L'angin,L. Sur le Septoria gr ami nun Deem., destructeur des leu- illes du ble. Compt. Kend. 126:1436-1440. 1698. '

i t

61 17. Marchal ,E. De la specialisation du parasitisme chez l'Erysi-

phe graminis. Compt. Rend. 135 ; 210-212. 1902.

16. Montemartini ,L. Sopra la specializzazione dei funghi parassiti con particolare riguardo alia specializzazione delle ruggini dei cereali. Riv. Patol. Veg. 8; 33-44; 145-158. 1916.

19. "ort ensen, M.L. Rostrup.S. und Ravn,P.K. Oversigt over Land- brugsplanternes Sygdome i 1910. Abstract in Centralbl f. Bakt. 33: 576. 1912.

20. MAller, K. Zur Biologie der Schwarzfleckenkrankheit der Ahorn- baume hervorgeruf en durch den Pilz Rhytisma acerinura. Centralbl. f. Bakt. 36: 67-98. 1912.

21. Neger,E. W. Beitrage zur Biologie der Erysipheen. ilora 90: 221- 272. 1902.

Norton, J.. 1.3, Host Limitations of Septoria lycopersici. Ab- stract in Phytopath. ±: 65. 1917.

23. Fands,P.D. Alternaria on Datura end Potato. Phytopath. 7: 327- 337. 1917.

24. Keed,G.M. Infection Experiments with Erysiphe graminis DC. Trans. Wis. Acad. 15: 135-162. 1905.

25. • Infection xperiments with the ^ildew on Cucurbits, Erysiphe ciohoracearum DC. Trans. Wis. Acad. 15: 527- 547. (1906; 1905.

26. The Mildew of the Cereals. Torrey Bot. Club. Bull. 36: 353-388. 1909.

27. • infection Experiments with the Powdery Mildew of Wheat. Phytopath. 2: 81-87. 1912.

28. • The Physiological Relations of the Powdery Mildews to their Hosts. Miss. Agr. Exp. Bull. 141. 1916.

29. • The Powdery Mildews of Avena and Triticum. Mo. Agr. Exp. Sta. Research Bull. 23. 1916.

30. Salmon,!!. S. On Specialization of Parasitism in the Erysiphaceae. Bot. Centralbl. 1,4: 261-315. 1903.

31. . Infection-Powers of Ascosoores in Erysiphaceae. Journ. Bot. 41: 159-165J 204-212. 1903.

32. . Cultural Experiments with the iiarley 'iildew, Erysi- phe graminis DC. Ann. Myc. 2: 70-99 1904

33. . Cultural Experiments with the Biologic Forms of the Erysiphaceae. Phil. Trans. Roy. 3oc. London, ser. E. 197: 107-122. 1904.

34. Salmon, E.S. On Erysiphe graminis DC, and its Adaptive Parasi- tism within the Genus Bromus. Ann. Myc. 2: 255-266; 307-343. 1904.

35. . Recent iiesearches on the Specialization of Para- sitism in the Erysiphaceae. New Phytol. 3: 55-60. 1904.

36. . On Specialization of Parasitism in the Erysiphaceae II. New Phytol. 3: 109-121. 1904.

37. . On Specialization of Parasitism in the Erysiohaceae III. Ann. Myc. 3: 172-184. 1905.

36. On the Stages of Development Reached by Certain Biologic Forms of Erysiphe in Cases of Non-infection. New Phytol. 4: 217-222. 1905.

39. . Further Cultural Experiments with 'Biologic norms' of Erysiphaceae. Ann. Bot. 19.: 125-148. 1905.

40. Shear, C.L. and Wood, Anna K. Studies of Fungous Parasites belonging to the Genus Glomerella. U.S. Dept. Agr. Bur. Plant Indus. Bull. 252. 1913.

41. Stager, R. Infect ionsversuche mit Graraineen-bewohnenden Clavi- ceps-Arten. Bot. Zeit. 61: 111-158. 1903.

42. Stakman.E.C. A Study in Cereal Rusts. Minn. Agr. Exp. Sta. Bull. 138. 1914.

43. Stakman,E.C. and Piemei 3el , F. J . Biologic ForraB of Puccinia graminis on Cereals and Grasses. Journ. Agr. Research. 10: 429-495. 1917.

44. Steiner.J.A. Die Speziali zation der Alchemillen-bewohnenden

Sphaerotheca humuli { DC J Burr. Centralbl. f. Bakt. 21: 677-736. 1908.

45. Stone, R.E. Studies in the Life Histories of Some Species of Septoria occurring on Ribes. Phytopath. 6; 419-427. 1916.

46. Ward.H.M. Further Observations on the Brown Rust of the Bromes, Puccinia dispersa lErikss.j, and its Adaptive Para- sitism. Ann. Myc. 1: 132-151. 1903.

47. rt'esterdijk, J . Unter suchungen ilber Sclerotinia libertiana als

Pflanzenparasit . Meded. Phytopathologi sch Lab. Willie Commelin Scholten 2: 1-26. Amsterdam. 1911.

Explanation of uraph I

This graph represents variations in Bpore length in 3eptoria verbascicola B.& C. and 3. tritici Desra. under a number of different conditions. The measurements of spores are indicated in microns by the base line, each space representing one micron. The frequency is indicated on the perpendicular lines, each space representing one spore. The measurements are at intervals of 2.4 microns, and for each curve 200 spores were measured.

A and B represent ranges in length of spores from a single

culture of j3. verbascicola upon onion agar; in Jl the spores were

from the lower portion of the colony which was moistened by the

small amount of water on the agar, while in B the spores were

from the upper edge of the colony where the agar was drying. C.

represents spores from spots on leaves of Verbascum blattaria in

the field where the light exposure was intense; B, spores from

shaded rosette leaves of the same host in the field; E, spores

from the same host kept under very humid conditions in the green-

house; F, spores of the same fungus growing upon Scrophularia

mar i land icq. , conditions as in E.

G represents the range of spore length of S« tritici upon

the upper stem leaves of naturally infected wheat plants in the

field in July; B, spores of the fungus from the same field taken

from the basal leaves of volunteer wheat in January. The plants

of wheat were dug up ana kept for a few days in a closed collecting

can in the greenhouse in the latter instance.

Explanation of Plate I

Fig. 1 Disease spots of Septoria polygonorum Desra. up-

on Polygonum penn sy Ivan i cum . Natural infection.

Fig* 2 disease spots of the same fungus upon P. persi -

caria . Natural infection.

Pig. 3 Disease spots of the same fungus upon P. lapath-

ifolium . Natural infection.

Pig. 4 Disease spots of the same fungus upon P. orientale ,

Artificial inoculation.

Fig. 5. Disease spots of S. malvicola E.& M. upon Malva

rotundi folia . Artificial inoculation.

Fig. 6 Disease spots of S. malvicola upon Althaea rosea .

Inoculated with spores from M. rotundifolia .

Fig. 7 Disease spots of S. lactucicola E.& M. upon Lac

tuce. scariola . Inoculated with spores from L. canadens i s.

Fig. 8 Disease spots of S. lactucicola upon L. sativa .

Inoculated with spores from L. canadensis .

Fig. 9 Disease spots of S. lactucicola upon L. canadensis

Natural infection.

Preparatory to photographing the leaves were treated with hot alcohol to remove the chlorophyll, but this process produced no apparent change in the character of the disease spots. The leaves were then softened in 50% glycerine and pressed.

The writer takes pleasure in making acknowledgement to

Dr. F.L. Stevens, Professor of Plant Pathology, University of Il- linois, for helpful suggestions and guidance in the preparation of this thesis. The writer also wishes to express his appreci- ation of the valuable assistance given by Prof. William Trelease,

Head of the Department of Botany, University of Illinois, in the solution of questions of taxonomy. Thanks are due Fred G. Seaver,

Curator at the New York Botanical Garden, and Vera K. Charles,

Mycologist, U.S. Dept. of Agriculture, for furnishing data upon the host ranges of various species of Septoria.

Biographical Sketch

Walter Spurgeon Beach was born near Brownton, Minnesota,

January 24, 1890. He attended the Hutchinson High School, Minn-

esota, and graduated in 1907. The three following years were

devoted to teaching in the rural schools of McLeod County of the

said state. In 1910 he entered the Minnesota Agricultural College

and in 1914 received the degree of Bachelor of Science in Agri-

culture. During the last two years of his college course he

taught as a student assistant in Botany. In 1914 he accepted a

position 88 a teaching fellow in plant pathology at the Michigan

Agricultural College and received the degree of Master of Science

in 1915. The Master's thesis entitled " The Fusarium Wilt of

the China Aster" was approved for a technical bulletin by the

Michigan Agricultural College. In the fall of 1915 he took a

position as Assistant in Botany at the University of Illinois,

and began study in the pursuit of the degree of Doctor of Phil-

osophy. In 1917 he was elected to the chapter of Sigma Xi at the

University of Illinois. In the college year 1917-1918 he was a fellow in Botany at the University of Illinois.